INDEX

Explanations

knowledgeable about, good, wants, work

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 вместо

-1.52

ПУ

-1.38

 instead

-1.30

ではなく

-1.27

 będzie

-1.19

коми

-1.16

 будет

-1.13

aterdag

-1.13

 cálculo

-1.11

SALUD

-1.11

POSITIVE LOGITS

and

1.63

 малы

1.33

during

1.28

 DURING

1.27

击败

1.26

even

1.23

 Genel

1.20

 même

1.20

周泽

1.20

higher

1.20

Activations Density 0.039%